OUP user menu

The effect of oxygen on the growth and cell morphology of Helicobacter pylori

Gianfranco Donelli, Paola Matarrese, Carla Fiorentini, Benedetto Dainelli, Tea Taraborelli, Emanuela Di Campli, Soraya Di Bartolomeo, Luigina Cellini
DOI: http://dx.doi.org/10.1111/j.1574-6968.1998.tb13248.x 9-15 First published online: 1 November 1998


The in vitro effect of progressive oxygen decrease on the growth and morphology of Helicobacter pylori was studied. H. pylori ATCC 43504 was used for the experiments. The strain inoculated in Brucella broth plus fetal calf serum was incubated under a controlled atmosphere with oxygen concentration from 5 to 0%. CFU ml−1 and bacterial morphology were detected at the time of spreading and at 24 h, 72 h, 7 days and 14 days. A detailed ultrastructural investigation of the bacterial cells, grown in different experimental conditions, was performed by scanning electron microscopy. Oxygen deprivation produced a rapid reduction of CFU ml−1. In particular, a significant reduction of viable bacteria was recorded at 72 h of incubation in the presence of 1% oxygen and anaerobiosis, and 0 CFU ml−1 was found after 7 days of incubation at the above mentioned oxygen concentrations. The coccoid phenotype was already prevalent after 24 h of incubation with a progressive tendency to aggregate in clusters. These clusters were progressively larger, depending on the reduction of oxygen concentration, since the aggregation phenomenon can be the expression of a hypothesized mechanism of protection among bacterial cells.

Key words
  • Helicobacter pylori
  • Oxygen
  • Morphology
  • Viable but not culturable state

1 Introduction

Helicobacter pylori is now defined as the main etiological agent of active gastritis and peptic ulcer [14] and the persistence of this germ in the stomach increases the risk of gastric cancer [5, 6].

While H. pylori infection is the most common gastrointestinal bacterial disease worldwide, knowledge of its mechanism of transmission is still poor [712].

H. pylori, as well as Vibrio-like bacteria, shows two different morphologic aspects: the typical spiral form and the coccoid form, the latter observed in vivo [13], and induced in vitro under stress conditions [1417]. The coccoid forms are not culturable in vitro and, for a long time, investigators have discussed their viability and their role in germ transmissibility and in infection relapse [13, 1723]. This morphology seems to be related to a specific form of ‘dormancy’ occurring in non-sporulating bacteria [15, 16, 24].

According to our data [25], coccoid forms represent the morphological expression of a viable but not culturable (VBNC) state, corresponding to a temporary adaptation to an unsuitable environment.

A focal point in the study of VBNC H. pylori is represented by the singling out of the different factors which can induce these morphological forms. Moreover, clear differences in density, DNA contents and secreted proteins among coccoid forms [26] are the consequence of various cultural conditions such as nutrient deprivation, antibiotic addition, unfavorable atmosphere.

The aim of this study was to evaluate, in vitro, the effect of oxygen deprivation on the growth of H. pylori and on the morphology of the bacterial cells.

2 Materials and methods

2.1 Bacterial strain and growth conditions

H. pylori ATCC 43504 was the standard control strain used for this work.

Microorganisms, stored at −70°C by the method of Drumm and Shermann [27], were defrosted to room temperature, and rapidly plated in chocolate agar plus 1% of IsoVitaleX (CA) (Becton Dickinson, Cockeysville, MD, USA) at 37°C for 3 days in a microaerophilic atmosphere (90% N2, 5% O2, 5% CO2, CampyPak Jar, Oxoid Ltd, Basingstoke, UK).

2.2 Determination of conversion rate under different oxygen concentrations

Bacteria harvested from almost confluent CA plates were inoculated in Brucella broth (BB, Biolife Italiana, Milan, Italy), supplemented with 2% fetal calf serum (FCS, Seromed, Biochrom KG, Leonorstiftung, Berlin, Germany), and incubated overnight at 37°C in a microaerophilic atmosphere with gentle agitation.

At the start of the experiment, the broth culture was divided into tubes of 14 ml each. Groups of four tubes were incubated in separate jars under the following condition of gas mixture: (a) 90% N2, 5% O2, 5% CO2 (standard condition), (b) 91% N2, 4% O2, 5% CO2, (c) 92% N2, 3% O2, 5% CO2, (d) 93% N2, 2% O2, 5% CO2, (e) 94% N2, 1% O2, 5% CO2, (f) 95% N2, 0% O2, 5% CO2 (anaerobiosis).

As control, experiments with either 10% CO2, 85% N2 and 5% O2 (A) or 10% CO2, 90% N2 and 0% O2 (B) were also performed.

Number of viable bacteria (CFU ml−1) and bacterial morphology were determined at time 0 and after 24 h, 72 h, 7 days, and 14 days.

2.3 Viable count

The number of viable bacteria was determined by plating 100 μl of 10-fold serial dilutions in BB plus 2% FCS broth cultures, on CA plates in controlled atmosphere (standard condition) at 37°C for 3 days.

For tubes incubated in atmospheres tending to anaerobiosis (2–0% of oxygen) and, for each mixture, after 7 days of incubation, the viable count was also performed by plating on CA in a controlled atmosphere at 37°C for 3 days, both 100 μl of 10-fold concentrated broth culture and the same aliquot of concentrated broth culture after passage in fresh broth (1 ml of broth culture plus 9 ml of BB+2% FCS and incubated overnight in standard microaerophilic environment at 37°C with gentle agitation).

This latter procedure was to evaluate the reactivation of bacteria either in coccoid form or aggregated in clusters.

2.4 Bacterial morphology

Bacterial features were studied in Gram modified stain with a Zeiss standard microscope at magnification 1000×. The morphologies, indicated as bacillary (B), or coccoid (C alone, or grouped in CL clusters) were counted as follows: for each broth studied, three slides were read and, for each slide, four fields were counted. Slides were read by two ‘blind’ microbiologists.

2.5 Electron microscopic observations

Ten milliliters of each broth culture were centrifuged for 20 min at 8000×g and were prepared for scanning electron microscopy (SEM) by fixing them with 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4) at 4°C for 30 min. After several washes in cold cacodylate, samples were resuspended in 40 μl of the same buffer. The solutions were dropped on 13-mm polylysine-covered coverslips at room temperature. Then, samples were postfixed in 1% OsO4 for 30 min, dehydrated through graded ethanols, critical point dried in CO2 and gold coated by sputting.

The samples were examined with a Cambridge 360 scanning electron microscope.

2.6 Statistical analysis

All experiment data are shown as mean±S.D. of triplicate determinations and the results are representative of two separate experiments.

Student's t-test (‘Statistic’ program for Macintosh) for correlated samples was used. A P value of less than 0.01 was considered significant.

3 Results and discussion

The growth of H. pylori ATCC 43504 cultures at different percentages of oxygen is shown in Fig. 1. Similar results were obtained with 5% and 4% oxygen concentration. Only at these oxygen tensions a significant increase of bacterial growth, compared to the starting concentration, was recorded (P<0.001). A peak of 5×108 CFU ml−1 was observed at 72 h of incubation and a significant reduction of CFU ml−1 until values of 106 CFU ml−1 was reached after 14 days (P<0.001).

Figure 1

Growth of H. pylori in cultures under different oxygen concentrations (a, 5% O2; b, 4% O2; c, 3% O2; d, 2% O2; e, 1% O2; f, 0% O2). Values are plotted as mean±S.D.

The variations of the morphological aspect of the germ were similar in the above considered environmental conditions (Fig. 2) with prevalence of the coccoid morphology after 24 h of incubation. Moreover, after 7 days of incubation, coccoid H. pylori tended to aggregate in clusters (Fig. 2a,b). These clusters were also observed in conditions c and d (Fig. 2c,d) in which constant values of CFU ml−1 were recorded (Fig. 1), with a reduced culturability in 2% oxygen (condition d). More stringent oxygen deprivation (1% and anaerobiosis, conditions e and f) revealed a reduction of 10-fold of CFU ml−1 at 24 h of incubation and after 7 days of incubation the loss of culturability (0 CFU ml−1) was significant compared to the previous readings (P<0.001).

Figure 2

Percentage of H. pylori forms detected under different oxygen concentrations (a, 5% O2; b, 4% O2; c, 3% O2; d, 2% O2; e, 1% O2; f, 0% O2). Each value represents an average of triplicate determinations and the results are representative of two separate experiments. The standard deviations range from −0.3 to +0.3. B, bacillary forms; C, coccoid forms; CL, clusters.

These latter two condition showed morphologically (Fig. 2e,f) a large presence of clusters starting from 24 h of incubation. In particular, at more stringent conditions, in which the percentage of oxygen was from 2% to 0%, an unusual morphology of H. pylori at scanning electron microscopy was detected. Long and undivided rods were observed after 7 days with a tendency to become filamentous (data not shown).

The general view of the changes observed in the morphology of H. pylori, grown in different oxygen conditions, is presented in Fig. 3. Panel A represents H. pylori cells grown at 4–5% oxygen concentration; panel B at 2–3% oxygen concentration; panel C bacteria grown at 1% oxygen and anaerobic condition.

Figure 3

Scanning electron micrographs of H. pylori in different oxygen conditions and growth times. A: 4–5% oxygen concentration; B: 2–3% oxygen concentration; C: 1% oxygen and anaerobiosis. Top line (Aa, Ba, Ca): 3 days of bacterial growth; middle line (Ab, Bb, Cb): 7 days of bacterial growth; lower line (Ac, Bc, Cc): 14 days of bacterial growth. Magnification: 2100×.

The micrographs on the upper (Aa, Ba, Ca), the middle (Ab, Bb, Cb) and the lower lines (Ca, Cb, Cc) show bacteria cultivated for 3, 7 and 14 days, respectively.

The most relevant aspect arising from the comparative analysis of the cells grown in different environmental conditions is represented by the increase in the number of clusters as a function of the time of bacterial culture. This phenomenon was more evident with the oxygen deprivation. In particular, in 1–0% oxygen, the presence of clusters (Fig. 2e,f) was earlier than in other oxygen conditions. Regarding the size of clusters, a maximum was reached when the incubation time was prolonged to 14 days and for the oxygen concentration of 1–0%.

Fig. 4 shows bacterial growth in both anaerobiosis and microaerophilic atmosphere obtained using 85% N2, 5% O2 and 10% CO2. In this microenvironment with a percentage of CO2 higher (10%) than that used for the above reported experiments (5%), CFU ml−1 decreased more rapidly and no viable bacteria were recorded after 14 days of incubation.

Figure 4

Growth of H. pylori in cultures detected in standard conditions (A) and in anerobiosis (B). Values are plotted as mean±S.D.

In conclusion, our results show that the presence of 5–4% oxygen in the gas mixture reveals a good growth of H. pylori, with an increase of 100-fold of the starting concentration after 72 h of incubation. The effect of oxygen decrease (3–2%) shows, after each control time, a relatively constant value of CFU ml−1. In fact, at 14 days of incubation bacterial growth was still observed at the above mentioned oxygen concentration. Further oxygen deprivation produced a rapid reduction of CFU ml−1. In particular, a significant reduction of viable bacteria (about two steps of concentration) was recorded at 72 h of incubation in the presence of 1% oxygen and anaerobiosis, and 0 CFU ml−1 was found after 7 days of incubation at the above mentioned oxygen concentrations.

Five instead of 10% CO2 enhanced the tolerance to the presence of toxic catabolic products and to the starvation.

Particularly interesting are the morphological changes observed at the ultrastructural level. Coccoid forms are more rapidly induced at low oxygen concentration with a tendency to aggregate with each other to form progressively larger clusters in accordance with recent data on adhesive properties of coccoid versus spiral forms of H. pylori[20].

Moreover, the presence of aggregate cocci can be interpreted as the modification of the behavioral answers of the germs which, in unfavorable conditions, tends to overcome the toxicity of the medium by their clustering.

Krieg and Hoffman [28] studied the effect of a high density of bacterial population on its own self-protection against the toxicity created in the medium because of increased oxygen; the more the broth culture is concentrated, the higher the protection of cells. An analogous mechanism could be effective in our opposite experimental conditions of oxygen deprivation in which the presence of clusters, more evident after 7 days of incubation, can express the tendency of cells to protect each other.


This work was supported by Grant CU.9704058.CT04 from the Consiglio Nazionale delle Ricerche and of the Ministero dell'Università e della Ricerca Scientifica e Tecnologica (60%).


View Abstract